What is possible with Google Ads API vs not possible with AdWords API?

[PPC Developer]

Hi, 

We've been building on top of AdWords API for 5 years now. Our integration has gone through solid time of stress testing and have a very stable server code base now. AdWords API has been very stable lately, as it's around 10 years old. This new Google Ads API is a massive undertaking, and I'm trying to understand what really is the benefit here? I read that it's using a different protocol and is strong typed, but the current API is working just fine. 

This whole migration will cost us so many man-months of coding and testing and stabilization afterwards. I would imageine hundreds of millions of dollars worth of pain and agony we have to go through collectively (globally, all Google API customers). And what do we exactly gain? 

1- There are no new scenarios or functionality enabled with the new APIs, am I missing something?

2- Some concepts and API services do not exist in new API, such as AdCustomizerFeedService, which seems that the new API is not fully ready yet.

3- There's a note about how the new gRPC protocol is implemented more efficiently than SOAP/XML. Well, the web as we know today is completely fine using SOAP/XML. Google's current APIs are running very well and fast. So what is exactly gained by switching to an entirely new protocol?

For us to fully migrate and stabilize the new code base, it will definitely take over 1 year of effort. I'm guessing there's more customers out there who share similar views. 

My question here for Google is : Why not keep supporting the previous AdWords API and keep publishing new versions? Looks like you guys changed the object model to be more in tune with the new AdWords UI. But why does that force you to stop supporting the AdWords API? Looks like both APIs have been tested to work side by side, so looks like a lot of work has been done from your end to support AdWords API in parallel with the new API. 

I'm trying to understand the rationale behind this massively costly industry-wide move, where it seems that there's no benefit or gain to us. Please help us understand how this new API is worth spending hundreds of millions of dollars from your partners point of view.

[Oliver]

Well said.  We still can't believe a company with the calibre, reputation and resources of Google can come up with such a comparably poor API for the exact one product that makes them most of their money.  We really are at loss as to what to make of this new API.

Oliver

[Ronak Shah]

Yes, I completely agree with this!

Adwords API has been very robust and stable since past many years and the new version is definitely one of the bad and worst (As per my my experience). The whole migration is not so easy, it would take so many months of work to map the old functions of Adwords API with the new Google Ads API.Plus I also feel the new version of API is slower and also poor API documentation and lacks some advanced code examples. And Finally I am surprised to know that Reporting CSV Download feature is removed, I am sure Reporting is the base and most used feature for most of the Tech companies and Downloading CSV and import data in database is very fast (Specially for big accounts) using old Adwords API, but with the new one its damn slower. For Example Lets say there is an Account with more than 5K Campaigns and we need to download the Campaign Performance Report, then from old Adwords API we easily import data from CSV, but now through new one since CSV Download is removed we need to iterate over each data which makes performance slower.

[Andre Tannus]

I'm just hanging back and watching this whole thing unfold for as long as I can.

I don't believe there are many casual users of the API. The changes it undergoes (api-breaking changes, deprecation of ad formats, new ad formats, etc...) mean you keep up or keep out.

A year may be enough to port an well written application (even one of meaningful size) to a new stable version, but what we are to port to is a "first stable version" of a ground-up rebuild, which means *unstable version* by any respectable quality criteria. Perhaps there was a beta. I never go an invite.

On that note, it bothers me profoundly that there is no clear warning that the current API it will be discontinued in less than 12 months. There is a subtle note in the Guide's "getting started" page which once you're started you'll probably never visit again and nothing in the Reference, which is where we (I at least) constantly go to when actively developing/maintaining api-related code. There should be a clear note on every page in the reference.

I have seen people asking questions about the current API that clearly indicate they are taking first steps into it. They get no warning that they are building against a sun-setting code base. Perhaps that is on them, since they should be reading the subtle note in the Guide.

All this makes me suspect "12 months after the first stable version" may be a soft deadline.

If not, we're all in for a ride.

Andre

[Mat]

Hi Andre,

there was a beta for a few months, but actually it wasn't even an alpha, since the features were far from the same as the AdWords API has (and they're still not on par).

The last "beta" version was v0_7. The next version was then given the version number v1 and the suffix "production ready" and we all see that it is clearly not. 

http://feeds.feedburner.com/blogspot/hecyUu

Regards

Mat